HL2DQ (Verified Cheating)


Hello everyone!

So, today I wanted to make an important (unrelated to us in anyway due to its age) announcement that we have wanted to talk about for some time, but felt our reputation would be harmed or cause an unnecessary backlash from previous runners.

First off, I want to express my gratitude to the runners of HL2DQ. Their hard work and dedication has brought our community together, and without it, we wouldn’t be where we are today, with 12,000 total forum posts, 10,000 subscribers and 1.7 million views on youtube, and over 300 members on our forums. We really can’t thank you guys enough. However, this does not include the runner, bandit5k.

Back in 2009, the new SourceRuns team discovered that HL2DQ was a cheated run, and should not count for any world record. This discovery is partially the reason DWaHMoV was made. Bandit5k, responsible for 13 segments throughout the run, decided he would take it upon himself to use both timescale, which is very clear in the Ravenholm segments, and, possibly, used different saves in some of the coast segments, which you can see when he has a different speed coming out of the tunnels. Because of his selfishness, he defiled what was an amazing achievement, and completely invalidated all of the runner’s hard work. I apologize tremendously to all the runners who participated in HL2DQ, especially DaleLewill, who has been extremely supportive of our community since we started.

Bandit5k has since tried making a comeback to speedrunning, as he joined the hype with Portal 2’s release back in 2011. Inexistence, TYPICAL, and I had to monitor his activity to make sure he would not ruin yet another run.

If anyone sees this person trying to participate in any speedrun-related project, please inform the run’s head. We do not stand for nor tolerate this kind of dishonesty, and we feel responsible for making this aware, as it does reflect on our community.

For further proof, please check this video out:
https://www.youtube.com/watch?v=ya9sXlnGctg (Demos slowed down 30% by Heplooner)
https://www.youtube.com/watch?v=bRtPi3s2uEU (A break-down analysis by BitRain)
https://www.youtube.com/watch?v=soFye1rP-zA (Comparison between Real-time and timescale by Centaur1um)

Thank you.


The said cheating is especially noticeable in d1_town_01a (and mostly onwards) which shows really twitchy and precise mouse movement, inhuman reflexes - something that’s only possible if done using host_timescale.


I’ve made a rough recreation of Bandit5k’s segment in host_timescale 0.3 to show how it would have been made and what it looks like in real time.

It’s the same, twitchy, jittery, super fast movements, probably too fast for a human, and very accurate


My whole world view is shattered :frowning:


It’s a shame, really. This project inspired most of us to run in the first place. But don’t be upset about all this.

It’s still a pretty sick run and a lot of the runners have shown their dedication to this project which is why both them and this run deserves to be respected regardless what. Most of the tricks discovered in this run are still being used in speedruns today and they still make up the majority of all tricks ever discovered for HL2. So if you can’t accept this as a legit speedrun anymore, try to see it as a collection of really good Half-Life 2 route demos that lead to the creation of this community and other great follow-up projects (such as DWaHMoV, lowerlogic’s crowbar-only run (really recommend watching it if you haven’t already) and so on…) ;D


Good job finding this out. It could be very important for future runs. =)

Im really angry right now when I remember when I watched this run the first time as a kid and was so happy about it… and EXACTLY this moment from bandit was the reason why I directly bought HL2. Because I thought I could be so fast too somewhen. This hurts me a bit now…

Anyway the run was nice…


And you’ve known this since 2009? :open_mouth: Of course, you had good reason not to come out with it any earlier; as you said it would hurt SR’s reputation as a new site to randomly accuse such a well known run of cheating. Though it is pretty alarming that some of us are just now finding out now how this inspirational speedrun that everyone (even new people like me who haven’t been interested in speedruns until recently) looked up to and considered an awesome piece of game-breakery. :stuck_out_tongue:

And perhaps the saddest thing is that so many people who genuinely had good, legitimate intentions for the project and worked on it for so long have had their work ruined by one guy who thought he could steal some segments with the help of host_timescale.

Though of course, as exec said, so many of the tricks that are used nowadays were found in the creation of HL2DQ, so there is a positive way to look at it :slight_smile: Just out of curiosity, roughly how many people did know about this before now?


I’m not sure. Not very many. As far to my knowledge, our reasoning as to not come out with this before was because Josh did not want to start drama. I think, I can’t be sure, and I won’t speak on his behalf. The reason I came out with it is because I believe differently than him, and believe it’s more important to call these things out to help enforce honesty within our own community, even if it does soil the achievement that brought us all together.

Runners from dwahmov figured it out themselves, so it just shows how obvious it is. If another group of people found out, it would directly taint our community, despite none of us having any direct involvement with hl2dq, aside from maybe Demonstrate.


That makes sense. And of course, someone would eventually figure it out anyway probably, given enough time. Plus, it doesn’t totally soil the achievement. It only soils a few minutes of the achievement. :stuck_out_tongue:

Edit: I wrote this before your edit so now it’s basically repeating what you said lol


And the whole point of speedruns to outsiders watching, which is the “wow look how fast they did this without cheating”


Really? I read some comments and nobody said that. :stuck_out_tongue:
That’s actually another frustrating thing about this. Normally people who are into speedrunning spend time explaining stuff to people who think runs are cheating, and now people who tried to correct people about HL2DQ are going to feel awkward. But oh well, generally YouTube comment trolls don’t care anyway so we shouldn’t worry about it too much I guess :wink:


So, heplooner has re-rendered the demos bandit recorded, and slowed them down by 30%. I suggest that everyone watches this, as it will help shed more light on this whole cheating thingy.



Wow, so it did have some cheated segments, huh. Funny thing is, as I watched the slowed-down breakdowns I couldn’t help thinking to myself, “yes it does look way too good to be true, but then again these guys are way too good to be true in general”. So yeah, it really doesn’t help that you guys are so much above us casuals, which is the reason not many people ever caught on to this instance in the first place. :stuck_out_tongue:

On a different note…

If anyone sees this person trying to participate in any speedrun-related project, please inform the run's head.

Pretty sure the announcement will make this impossible, since as long as bandit5k sees it he’ll stop using that handle for anything speedrun-related. :stuck_out_tongue:

Edit: Heh, registered in 2010 and still labeled as a new member.


It is upsetting to hear of this. I remember seeing some of these segments and thinking how fast they were. Back then, I wasn’t involved in doing any of the segments, but I liked helping out the team when I could with testing route possibilities. HL2DQ definitely got me interesting in doing speedruns in the Source engine, which ultimately lead me to do PDP.

As we are getting closer to Source 2 (and ultimately, Half-Life 3), it is good to make sure that any project that takes place on this site is verified accurately. Back during HL2DQ, I wasn’t even aware of all of the capabilities of the Source engine. For me, I only began to use time-scale for testing Portal segments and tricks. It is fine to use these built-in tools to figure things out, but using them and claiming you performed those actions legitimately is simply wrong.

I hope we can continue to keep the integrity of SourceRuns for future Source 2 game runs, and our continued GoldSource and Source runs.


I think Half-Life 3 won’t be released. Except Valve has money problems…
so yea… never.

But Source 2 will be awesome I guess! Even tho there won’t be a way to bhop I guess… but maybe we’ll find an other bug then. :stuck_out_tongue:


There are Half-Life 3 references everywhere in Valve games/Source engine codes, it’s been confirmed for a long while that they’re working on it.


And what on Earth do you mean Valve has money problems?


Pretty sure he meant they won’t release HL3 UNLESS they fall into money trouble - in which case releasing the game, even if terrible and half-finished, will make them millions and prop them back into the black. As long as they have money, “perfection will remain the enemy of good” and we won’t see the game.


So they’re waiting till they have no money to make more money?! That’s like the exact opposite of every other business model I’ve ever seen. Nah, that seems absurd for a large corporation.


That’s still not really the idea. We currently have two (assumed) factors:

  1. Valve has a lot of revenue streams (Steam, trading DOTA/CS:GO etc)
  2. Valve wants HL3 to be mindblowingly awesome and amazing.

However, reaching ‘perfection’ is really really hard, especially in an industry which is as fast moving as the current one. Duke Nukem Forever and Daikatana are examples of two games that tried constantly to be ahead of the curve, but kept on finding themselves behind it. They have a decent game working, but another game comes out with weather effects, so you think: we need that in our game! Your game gets delayed, and in that time, another game comes out with cool physics or something.

That may not even be the case, it might be self-imposed perfection. “I want these weather effects to be even better than they are now!”

Gooseman / Minh Lee, the creator of Counter Strike said he left Valve because he had *too much *freedom. Essentially no limit to funds, no boss and no time limits. He was unable to push anything out in those conditions and left to form a company with a different structure. Now that he had to build something or starve to death he could come out with Tactical Intervention.

If Valve suddenly found themselves in monetary issues this year, they suddenly have a *need *to increase revenue and may very well release HL3 next year. As is, they are perfectly capable of waiting and trying to make the game as good as possible.

So its not about them waiting to have little money; its about them waiting for the game to be perfect until they release it. If they have money problems, they’ll have to release it even if it isn’t perfect in their eyes.

This is getting way off topic now, sorry.